Convenient and objective assessment of gait is important for health management of older adults. Measuring gait function and mobility objectively can assess adherence and progress in an exercise or rehabilitation program, or track dosage and efficacy of medication. Accurate measurements can be taken in the clinic or laboratory, but field-based measurements would greatly assist health management. Unfortunately, there is little means of accurately performing real-world, long-term assessments in the home and at reasonable cost. Miniature sensors are rapidly improving in accuracy and power economy, offering great potential for field-based activity assessment. Simple accelerometer-based pedometers have proven highly accurate for step counts, and miniaturized, wireless systems such as the Nike+Ipod Sport Kit provide estimates of running speed and distance accurate enough for casual use, while remaining unobtrusive enough to wear conveniently for long durations. The requirements of research or clinical mobility assessments, however, exceed those of current systems. Several convergent technologies make highly accurate, miniature, inertial sensor systems feasible, including multi-axis accelerometer and angular rate sensors in small chip packages, powerful microprocessors with built-in wireless transmission capabilities, and increases in memory density. Most importantly, model-based sensor integration technology makes it possible to fuse data from multiple sensor types and minimize drift error. The aims of this project are to: 1. Develop a Field-Based Gait Assessment System, which will integrate information from miniature inertial sensors attached to the body and limbs, and process these data onboard to quantify walking speed, stride length, step variability, and other gait variables. We will design the sensor packages mounted in shoes to transmit wirelessly to integration and processing unit mounted at the waist. [unreadable] 2. Test and calibrate this system on human subjects, using simultaneous and independent laboratory-based measurements. We will use optical motion capture to assess actual gait variables using standard techniques of known accuracy. This project seeks to evaluate the gait of elderly individuals to assess fall risk, aid fall prevention, promote exercise, diagnose disease progression, and evaluate medication. This project will develop a set of miniature, wearable, wireless sensors that can record gait activity over long periods of time (> 1 day) with minimal interference with daily activity. [unreadable] [unreadable]